The present invention is founded on electromagnetic-field scalar projection principles and optimal matched filter concepts that define a hybrid field detection process capable of coherently imaging millimeter radio-frequency (mmRF) through Ultra-Violet (UV) optical fields without optical filtering, scanning, or use of other moving components, while at the same time preserving the simplicity of conventional, direct-detection hardware.
The physics governing the conveyance of spatial information with electromagnetic fields is described by the well know Huygens-Fresnel principle. In essence, the principle treats the scattering of scalar electromagnetic fields, e.g., from an extended target's surface, as the superposition of interfering wavelets emanating from scattering centers distributed over the target. Conveyance of the resulting spatial information from a target to a detector is through transverse co-phasal surfaces oriented normal to the longitudinal direction of propagation.
Referring to FIG. 1, spatial amplitude and phase distributions comprising the transverse field are contained in the two-dimensional surface (x, y) normal to the longitudinal direction of propagation (z) with constant phase surfaces separated by a wavelength. The propagating object field conveys spatio-temporal image information to an observer or detector at z via temporal variations in the spatial amplitude and phase.
Observation of the spatio-temporal information will, in one form or another, ultimately reduce to a power measurement that requires extraction of energy from the propagating field. The basic field detection processes are generally categorized as either coherent or incoherent.
Incoherent detection subjects the field to a process referred to as squared modulus or square-law detection, for example, when imaging visible light with eyes, film, CCD camera or other photosensitive device. Ideal square-law detection of a complex scalar field consisting of the spatial amplitude and phase distribution, ψ(x,y,z,t), results in the following intensity distributionI(x,y,t)∝|ψ(x,y,z,t)|2≡ψ(x,y,z,t)·ψ(x,y,z,t)*,  (1)where “*” is the complex conjugation operator. Note that all phase information is lost in the detection process.
A coherent detection process endeavors to preserve phase information via the interferometric (homodyne or heterodyne) mixing of the object (image) field, ψ(x,y,z,t), with a known reference field, ψRef(x,y,z,t), followed by square-law detection and filtering, such that the interfering cross term is extracted
                                                                                    ψ                ⁡                                  (                                      x                    ,                    y                    ,                    z                    ,                    t                                    )                                            +                                                ψ                  Ref                                ⁡                                  (                                      x                    ,                    y                    ,                    z                    ,                    t                                    )                                                                          2                ⁢                  ⇒          Filter                ⁢                              ψ            ⁡                          (                              x                ,                y                ,                z                ,                t                            )                                ·                                                                      ψ                  Ref                                ⁡                                  (                                      x                    ,                    y                    ,                    z                    ,                    t                                    )                                            *                        .                                              (        2        )            
The fundamental advantages of coherent over non-coherent detection processes include the ability of filterless separation of object field from clutter signals and background, controlled noise bandwidths, and coherent amplification. Coherent imaging will enable system designers to exploit a broad range of coherent detection processes including phase/frequency discrimination, Doppler, range, three-dimensional imaging, synthetic aperture imaging, speckle manipulation, and multi-wavelength spectroscopy and interferometry.
The highly idealized coherent detection process implied by Equation 2 is extremely difficult to achieve at shorter mm-RF through optical wavelengths in practice. This is because the implementation of Equation 2 for coherent imaging is hindered by the intrinsic difficulty in matching the spatial and temporal field components of object and reference fields at shorter wavelengths. In essence, the image- and reference-field's time-varying spatial amplitude, phase, and polarization distribution (FIG. 1) become more difficult to match when the transverse dimensions of the receiver aperture are large when compared with the field's wavelength.
As detailed in U.S. Pat. No. 5,875,030, Method and Apparatus for Coherent Electromagnetic Field Imaging Through Fourier Transform Heterodyne, issued on Feb. 23, 1999, to Cooke et al., the spatio-temporal matching of object field and reference field is essential for coherent field imaging. Coherent field imaging can be perceived as the scalar projection of the reference field onto the object field with the intermediate-frequency signal current conveying the amplitude and phase of the projection. In other words, only image field and reference field states that are spatio-temporal matched will contribute to image formation. Hence, a stable coherent imaging process requires the precise temporal and spatial matching of object field and reference field states. Hence, the progressive loss in spatio-temporal coherence between object and reference fields will eventually lead to the systematic phase and amplitude degradations in the detection process, resulting in the loss of conversion efficiency, corrupted image retrieval, and increased noise bandwidth.
For these reasons, state of the art in coherent imaging must address real environments where spatio-temporal mismatches are introduced by a variety of means including transceiver source drift and fluctuations, atmospheric turbulence, scattering from extended (rough) targets, and target motion. For example, unavoidable temporal mismatches between image and reference fields resulting from the finite coherence length of the laser source, and target induced Doppler shifts necessitate a receiver with large detection bandwidths (˜107-108 Hz). This prohibits the use of readily available focal plane array or CCD imaging technology, which are limited to 1000 Hz or less. The limitations imposed by large bandwidths are currently overcome by building up an image through synchronous target scanning, i.e., the rastering of a narrow laser beam across a target and building the image one sample element per unit time. With synchronous scanning, detection elements in the receiver may be limited to a single detector or to small (2×2-4×4) arrays. However, the resulting scanned image is not necessarily spatially coherent, or correlated, as the image is constructed from effectively independent samples.
Assuming the availability of high-bandwidth large-format imaging arrays, present heterodyne techniques would suffer from relatively poor and unstable detection efficiencies. Spatial mismatches between image and reference fields caused by source, atmosphere, and target-induced field fluctuations and distortion result in low coherent detection efficiencies. Compounding the low detection efficiencies are the superimposed scintillation-induced fluctuations. Scintillation is the result of time varying constructive and destructive phase fluctuations between image and reference fields and can completely dominate the image signal-to-noise ratio (SNR) statistics.
U.S. Pat. No. 5,875,030 addresses the spatial mismatches inherent in coherent imaging. The method decomposes the spatial components of the image field in terms of spatial reference functions. Selection of appropriate spatial basis functions allows image retrieval by way of classic Fourier manipulations enabling the direct imaging of the transverse amplitude, phase, and Doppler shift of coherent electromagnetic fields. However, one limitation with Fourier Transform Heterodyne (FTH) is that the image must remain stationary during the time required to build up the Fourier coefficients. This is because FTH employs a single-element detector and requires the serial scanning of basis functions for acquisition of one image coefficient at a time. Any movement or change in the target image during this period will cause an error during image retrieval.
The present invention builds on scalar projection and FTH principles, thereby enabling practical coherent imaging systems based on conventional focal plane array or CCD imaging technology, without the serial scanning of basis functions required in FTH. The present invention permits the engineering of electromagnetic fields for optimal coherent field imaging at short wavelengths. These so-called hybrid fields provide the mechanism through which the spatial and temporal matched filter conditions required for short-wavelength coherent detection are realized.
A hybrid field is the superposition of two or more matched fields, with coherent detection uniqueness introduced through the selective spatial or temporal modulation of one field relative to its identical match. Modulation formats include amplitude, phase, frequency and polarization, and are necessary because they create a uniquely detectable interferometric signature under square-law detection.
In contrast to the current practice of coherently mixing the return field, the hybrid is coherently mixed (interfered) before leaving the transmitter source, e.g., reducing transceiver and environmental spatio-temporal mismatches. Furthermore, hybrid mixing transforms the coherent detection process from one proportional to the (short) carrier wavelength, λo [m], to one proportional to a longer synthetic wavelength, λs [m]. This transformation applies to all interferometric processes, including object-to-reference field coherence, sampling requirements, spatial resolution, and range/Doppler. Hence the ability to mix before transmission, and the generation of controlled synthetic wavelengths permits efficient coherent detection of range, Doppler, and multi-channel imaging with conventional focal plane array or CCD imaging technology. This precise control over the hybrid field enables achieving the spatial and temporal matched filter condition, e.g., the precise temporal and spatial matching of all object and reference field states, for error free imaging.
Various objects, advantages and novel features of the invention will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following or may be learned by practice of the invention. The objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.